GSoC 2007 Progress?

What is the current progress on the GSoC 2007 blender projects as the wiki is not being updated and i can’t find anything in the forums? Has all the projects been finalised and when can we expect to see a release of the GSoc projects.

Regards and thank you to all the coders burning the midnight fires to make blender the best.

Jacques

GSOC ended months ago. Most of them are working toward integration now.

Martin

Which project(s) are not going to see the light of day?

Render API is probably dead.

Brecht (one of the coders @ Peach) said at the bf_commiters mailing-list that he would continue to work on making a Render API, but he’s all for using his own method (which is understandable, even though I’m pro-Render API because it makes it possible to get data from Blender instead of getting it spoonfed directly… more flexible imho).

Deep shadow maps is not dead and still being worked on, but the coder has a bunch of nightmares about some bugs.

Mesh bevel and offset tools/modifiers depends on bmesh, afaik.

Dunno about what happened to the audio system cleanup/upgrade and the GLSL display mode. If I recall correctly, the audio system was almost done, but not sure. You’d have to check the mailing lists yourself if you’d like to know :slight_smile:

GSOC graveyard:
Sketch based modeling
Realistic skies and clouds
Nurbana (actually, is in ‘undead’ mode I think)
Pytextures
ODE physics integration in animation
Nurbana
Verse?
Mesh bevel
Render API
GLSL?

I think that most of the GSOC failed because the projects were too ambitious.

There were 3 ambitious projects that compensated (and that is an understatement) the failed projects more than enough:

  1. Fluid Sim
  2. Sculpting
  3. Animation rewrite

The verse code that was done for GSOC has been in Blender for a long time. It has to be enabled at compile time (not sure if it’s done for official builds).

Mesh Bevel, as mentioned earlier in the thread depends on BMesh and has been merged in the BMesh branch (IIRC).

Don’t know for GLSL, but the whole thing was successfull, if memory serves right, and is only waiting integration.

Martin

Mesh bevel can probably be integrated directly into trunk with a couple hours of work. It does depend on bmesh, but bmesh is very much a standalone module so it can work alongside blenders current mesh structures.

I will have to talk to Levi about it, but you can probably expect bevel to be in next official release.

Cheers,
Briggs

Do you mean Pynodes?
If yes, Jesterking has a SVN branch for Pynodes and he keeps it up-to-date with the trunk (merging changes every weeks) which probably means he plans to merge it soon (or keep it in a mergeable state).

Pytextures were made obsolete by pynodes, and I see pynodes in the SVN archives being merged forth and back often.

Sketch based modeling - the author quit part way through as I recall, and I don’t think he recieved the remainder of the funding

Realistic skies and clouds - work was completed but artist feedback was not favorable for the results so Ton decided against integrating it

Nurbana (actually, is in ‘undead’ mode I think) - the author got busy with other things, but has returned to it in recent times.

Pytextures - was replaced by pyNode which will probably be in the next release

ODE physics integration in animation - the author never started it and never recieved any google payment AFAIK

Nurbana - repeat :slight_smile:

Verse? - was integrated but still is not ‘feature complete’ and stable enough for general usage.

Mesh bevel - will probably be in next release

Render API - it was determined that the scope was too large and overly ambitious. It gave us good research on what issues need to be overcome - it won’t be integrated

GLSL? - the author is currently busy but plans to pick it back up in the future as mentioned on bf-committers

LetterRip

Thanks for clearing up any issues LetterRip. At least now we know that
not all of the 2007 GSoC projects are dead! :cool: